Microscopy:
Segregation patterns for pollen tetrads (Fig. 3c) were analyzed using an epi-fluorescence Nikon Eclipse 1000 microscope (Melville, NY, USA) equipped with filters from Chroma Technology (Rochester, NY, USA).
Methodology:
In the F2 and F3 generations, plants which were hemizygous for I3a interval and homozygote mutant for qrt1-2-/- were selected for tetrad analysis. Small drops of PGM media (34 % sucrose, 4mM CaCl2, 3.25 mM boric acid, 0.1% Triton-X pH7.5) were added to a slide and one flower was dipped into the drop, with its anthers pointing down, to collect pollen in the PGM solution. The DsRED and CFP filters, or a multi-color filter, were used to find plants that produced tetrads (qrt1-2) with a 2:2 segregation pattern (Fig. 3c). Pollen tetrads were divided into three categories: PD, NPD, and TT (Fig. 3c), and data
RESULTS:
Genotyping results for the AtGEN1-1 and AtGEN1-2 alleles are shown in Fig. 6. Wild type primers were designed to flank the insertion site, this way if the insertion is present, these primers won’t amplify under standard PCR conditions. Confirmation of the presence of the insertion was also checked by using a primer that starts from within the T-DNA insertion in combination with one of the wild type primers (Table 2). Combining results obtained from these two sets of reactions we are able to determine whether a plant is wild type, heterozygote or homozygote mutant for the desired T-DNA insertions. For AtGEN1-1 at first the 5′ GEN1-1#1 and 3′ GEN1-1#1 primers were used, producing PCR products of the expected size for both the wild type sequence and T-DNA insertion site using the 5′ GEN-1#1 and 3′ Lbb1.3 combination (Table 2.) (Fig. 6) (Appendix I). Confirmation of the T-DNA insertion in this region was made by sequencing a purified PCR product from an agarose gel (Appendix I). The 5′ GEN1-1SALK and 3′ GEN1-1SALK confirmed Salk T-DNA primers were later ordered to improve upon the PCR reaction; they were verified by amplifying controls previously used and produced more robust bands, thus were used for subsequent PCR reactions (Fig. 6). The insertion was found to be 39bp upstream of the ATG start site in the 5′ UTR (Appendix).
For the genotyping of the AtGEN1-2 allele the combination of 5′ GEN1- 2SALK and 3′ Lbb1.3 generated the correct predicted PCR product size based
on the site of the insertion, and yet sequencing revealed a number of anomalies. The combination of the wild type primers 5′ GEN1-2SALK and 3′ GEN1-1SALK produced the right product (Appendix I). Analysis of the 400-500bp fragment produced from the T-DNA reaction (Fig. 6), revealed overlapping sequences using both sets of primers (Appendix I). As is the case with many T-DNA
insertion lines, it’s possible that duplication(s) of the LB site has occurred leading to multiple priming sites within the insertion leading to multiple similarly sized fragments being amplified. Also, in the Frequently Asked Questions page for the Salk Institute T-DNA lines (http://signal.salk.edu/tdna_FAQs.html), other
researchers have observed the LBb1 primer alone can produce a band of ~450bp even in WT DNA under certain conditions (Question 25) and it was explained that sometimes it can appear as an artifact of this particular primer. Therefore it may be this artificial band that is creating the seemingly overlapping sequences. Because of the size of the wild type band (~1200bp) false positives can be avoided.
Interestingly, when analyzing the sequence produced using the 5′ GEN1- 2SALK primer, overlapping sequences were observed, and yet using BLASTn the predicted sequence produced was located to At4G39950. Because of the poor quality of the sequencing data, it is hard to assess whether this is a true second insertion site or not, especially since sequencing the same band using the 3′ Lbb1.3 primer produced such poor results from overlapping bands that the
sequence prediction lead to no BLAST result hits (Appendix I). Also, in a previous sequencing attempt no prediction could be made from the resulting predicted sequences also because of the existence of too many overlapping sequences. The sequencing results are particularly perplexing due to the fact that these primers were obtained from the source that confirmed the location of the insertion as being homozygote (http://signal.salk.edu/tdna_FAQs.html), and when using both forward and reverse wild type primers the correct sequence is produced. Combining these data it seems possible that multiple LB sites in the insertion exist, which are producing confounding multiple sequences. Due to this ambiguity, the data produced using the AtGEN1-2 x I3a allele needs further verification through methods discussed later.
Two datasets were analyzed for the AtGEN1-1 x I3a and one for AtGEN1- 2 x I3a as described in the methods. Linkage analysis was performed using the Stahl Lab Online Tools application “Map Distance, Interference, and Statistical Significance Based on Tetrad Data - The Traditional Method” program
(http://molbio.uoregon.edu/~fstahl/). This method utilizes the Perkins equation (Burgeff, 1929; Perkins, 1949), and the Papazian formula (Papazian, 1952) to calculate the genetic distance between the two markers in interval I3a. The data entered into the analysis program is displayed in Tables 4 and 5 and the results are summarized graphically in Figures 7 and 8.
Fig. 6 – Gel Electrophoresis (1.2% Agarose, TE Buffer) of PCR products. A. The expected WT band is ~1100 bp, while the band from the T-DNA insertion is ~500 bp. Presence of both
indicates a heterozygote plant. Control reactions using WT and MT DNA do not amplify using the MT and WT primers respectively. B. The expected WT band is ~1200 bp, while the band from the T-DNA insertion is ~400bp. Presence of both indicates a heterozygote plant. Control reactions using WT and MT DNA do not amplify when using the MT and WT primers respectively.
CO Analysis of AtGEN1-1 and AtGEN1-2
Gen1-1 x I3a (Set#1)
PD NPD TT TOTAL cM S.E.
WT 638 6 606 1250 25.68 0.8875
HET 3044 32 2744 5820 25.22 0.4222
MT 1919 25 1979 3923 27.13 0.5291
Total= 10993
Table 3. Tetrad counting data and Analysis for Set#1. PD (Parental Ditype), NPD (Non-Parental Ditype), TT (Tetratype), cM (Centimorgans = genetic distance), S.E. (Standard Error), WT (Wild Type - Gen1-1+/+; I3a+/-), HET (Heterozygote - Gen1-1+/-; I3a+/-),), MT (Mutant – gen1-1+/+; I3a+/-)
Gen1-1 x I3a (Set#2) PD NPD TT TOTAL cM S.E. WT 1073 7 834 1914 22.88 0.6837 HET 6456 87 4986 11529 23.89 0.3212 MT 5010 67 4013 9090 24.28 0.3599 Total= 22533
Table 4. Tetrad counting data and Analysis for Set#2. PD (Parental Ditype), NPD (Non-Parental Ditype), TT (Tetratype), cM (Centimorgans = genetic distance), S.E. (Standard Error), WT (Wild Type - Gen1-1+/+; I3a+/-), HET (Heterozygote - Gen1-1+/-; I3a+/-),), MT (Mutant – gen1-1+/+; I3a+/-)
Fig. 7. Genetic distances between siblings segregating for the gen1-1-/-mutation. Standard Error
bar details in Tables 2 and 3.
In the first dataset analyzed for the T-DNA insertion allele AtGEN1-1 x I3a a non-statistically significant increase in the genetic distance was observed between wild type and mutant genotypes (25.68cM vs. 27.13cM), the same pattern was observed in the second set analyzed (22.88cM vs, 24.28cM) (Table 4, Fig. 7). Significance is calculated by two methods: method 1 by looking at the variance of the two genetic distances and the standard error, if they don’t overlap
the difference might be real, while the method 2 is by looking at the absolute value of the differences between the two map distances, if the difference is more than twice the standard error, it is thought to be significant
(http://molbio.uoregon.edu/~fstahl/EquationsMapDistance.html). For both datasets method 1 lead to a “maybe significant” outcome, meaning that taking into account the standard error of each, the genetic distances did not overlap. Method 2 on the other hand showed that this difference is less than twice the standard error; therefore the net difference between the genetic distances of wild type and mutant genotypes is not-significant.
Gen1-2 x I3a PD NPD TT TOTAL cM S.E. WT 397 3 310 710 23.1 1.1496 HET 1574 27 1337 2938 25.51 0.6688 MT 392 6 385 783 26.88 1.236 Total= 4431
Table 5. Tetrad counting data and Analysis. PD (Parental Ditype), NPD (Non-Parental Ditype), TT (Tetratype), cM (Centimorgans = genetic distance), S.E. (Standard Error), WT (Wild Type - Gen1- 2+/+; I3a+/-), HET (Heterozygote - Gen1-2+/-; I3a+/-),), MT (Mutant – gen1-2+/+; I3a+/-)
Fig. 8. Genetic distances between siblings segregating for the Atgen1-2-/-mutation. Standard Error bar details in Table 4.
In the dataset analyzed for the putative T-DNA insertion allele AtGEN1-2 x I3a a statistically significant increase in genetic distances was seen between wild type and mutant genotypes (23.1cM vs. 26.88cM) (Table 5, Fig. 8). By the same methodology presented for AtGEN1-1 x I3a, method 1 showed that the
differences might be real, and by method 2 it was confirmed to be signigican by showing that the absolute value of the difference is more than twice the standard error. .